FOR ONLINE PUBLICATION ONLY

Appendix A

Information on the sources and calculations for the estimated carbon fluxes due to animal presence presented in Table 1.

Estimation of Arctic grazing herbivore effects

Zimov and others (2006) state that Yedoma permafrost covers 106 km2. Using laboratory incubations of thawed soil to measure long term (several year) respiratory release of CO2, they found that soils emitted 0.5 to 5 g C m3 d-1. Assuming a 1 m deep active layer (Zimov and others 2006), we estimate that Yedoma could potentially release about 183 -1825 Tg C over the entire 106 km2 Yedoma region.

Estimation of Moose effects

Data used to estimate moose contributions to whole ecosystem carbon cycling were obtained from experimental research conducted in Isle Royale Michigan, USA. Moose influence ecosystem carbon uptake through browsing, by priming soil microbial activity through fecal deposition, and through direct respiration (Pastor and others 1992).

Net primary productivity (NPP) of trees in areas excluding moose amounts to 8000 kg C ha-1 y-1 or 8 x108 g C km-2 y-1 (McInnes and others 1992). Comparisons of impacts of moose at varying densities also reveals that NPP declines by between 700 and 1500 kg C ha-1 y-1 per unit of moose (McInnes and others 1992). Therefore, under higher moose density (~ 1.5 moose km-2), NPP ranges between 5.8 x108 – 6.95 x108 g C km-2 y-1, and under low density (~ 0.5 moose km-2), NPP ranges between 7.25 x108 – 7.65 x108 g C km-2 y-1. (These moose densities fall at the low end of the density range of 0.4 to 9.3 moose km2 reported for the boreal forest globally [Feldhamer and others 2003]). The boreal forest ecosystem in Canada covers an area of 2.54 x 106 km2, of which 1.89 x 106 km2 is 90-100% intact (Canadian Forest Service 2005. Using this conservative estimate of intact forest, and assuming wood biomass contains 50% carbon (Jain and others 2010), we estimate boreal NPP in wood to be 756 Tg C y-1. This value compares favourably with an independent estimate of 809 Tg C y-1 across Canada based on forest inventory data from managed forests (Stinson and others 2011).

We estimated that NPP for the 1.89 x 106 km2 intact boreal in Canada to be 548 Tg C y-1 – 656 Tg C y-1 under higher moose densities and 689 Tg C y-1 – 722 Tg C y-1 under low densities. This leads to a difference of ~ 66 Tg C y-1 – 141 Tg C y-1. This estimate of carbon uptake does not account for potential carbon release through heterotrophic respiration (moose and microbes). Soil respiration is estimated to be 2.54 x 104 g C km-2 y-1 under low moose density, and rise to 3.43 x 104 g C km-2 yr-1 (Pastor and others 1993). Moose respiration varies seasonally being higher in spring, summer and fall than in winter (Regelin and others 1985). We used spring, summer and fall respiration as a liberal measure of C release. Moose metabolic rate during this time is 552 KJ kg-0.75 (Regelin and others 1985). Assuming an individual moose has a body mass of 359 kg (Regelin and others 1985), we estimate a respiration rate of 4.55 x 104 KJ individual-1 d-1, or 1.1x103 g C individual-1 d-1, assuming 41 KJ (g C)-1. Under low moose densities on Isle Royale, annual (365 d) respiratory release of C is 3.73 x105 g C km-2 yr-1 and rises to 1.73 x106 g C km-2 y-1 under high moose densities. Collectively, soil and moose respiration is orders of magnitude less than tree NPP. Thus, effects of moose on carbon exchange is dominated by tree NPP.

Estimation of Mountain Pine Beetle effects

Estimates are based on simulations of pine beetle effects in a 75,000 km2 area within interior British Columbia, Canada that was subject to a pine beetle outbreak (Kurz and others 2008). Average annual emissions were estimated from data presented in Figure 4 (Kurz and others 2008) for the 17-year period after the 2003 outbreak for the control scenario (low beetle abundance) and the beetle scenario (beetle outbreak).

Estimation of Wildebeest effects

Estimates of carbon flux for the 25,000 km2 Serengeti savanna ecosystem under historically low wildebeest density (250,000 animals in 1960) and currently high wildebeest density (1,250,000 animals in 1990’s) were obtained from Figure 5B in Holdo and others (2009). These values do not account for direct release from the animals themselves. Based on methane release in cattle (in CO2 equivalents [Dlugokencky and others 2011]), we estimated that per capita release for a 100 kg individual would be ~2.75 x 105 g C y-1. Emissions directly from animals (when per capita effects were scaled to low and high population densities: 6.8 x 1010 – 3.3 x 1011 g C y-1) were 1-2 orders of magnitude less than the whole ecosystem level effects.

Estimation of marine fish and marine echinoderm effects

Estimates of marine fish CaCO3 production were obtained from Figure 1 in Wilson and others (2009). Estimates of marine echinoderm CaCO3 production were obtained from Table 5 in Lebrato and others (2010).

Estimation of Sea Otter effects

Estimates of sea otter contributions to marine carbon flux were obtained by scaling per m2 estimates of net primary productivity presented in Table 1 in Wilmers and others (2012) to the entire 12,000 km2 region occupied by sea otters.

References

Canadian Forest Service. State of the Canada’s forests 2004-2005: Boreal forest. Government of Canada Publications (2005).

Dlugokencky EJ, Nisbet EG, Fisher R, Lowry D. 2011. Global atmospheric methane: budget changes and dangers. Phil T R Soc A 369: 2058-72.

Feldhamer GA, Thompson BC, Chapman JA (eds). 2003. Wild Mammals of North America: Biology, Management and Conservation (Baltimore: Johns Hopkins University Press).

Jain TB, Graham RT, Adams D. 2010. Carbon concentrations and carbon pool distributions in dry, moist, and cold mid-aged forests of the Rocky Mountains. USDA Forest Service Proceedings RMRS-P-61.

Holdo RM, Sinclair ARE, Dobson AP, Metger KL, Bolker BM, Ritchie ME, Holt RD. 2009. A disease-mediated trophic cascade in the Serengeti and its implications for ecosystem C. PLoS Biol 7:e1000210.

Kurz WA, Dymond CC, Stinson G, Rampley GJ, Neilson ET, Carroll AL, Ebata T, Safranyik L. 2008a. Mountain pine beetle and forest carbon feedback to climate change Nature 452:987-90.

Lebrato M, Iglesias-Rodriguez D, Feely RA, Greeley D, Jones DOB, Suarez-Bosche N, Lampitt RS, Cartes JE, Green DRH, Baker B. 2010. Global contribution of echinoderms to the marine carbon cycle: CaCO3 budget and benthic compartments. Ecol Monogr 80:441-67.

McInnes PF, Naiman RJ, Pastor J, CohenY. 1992. Effects of moose browing on vegetation and litter of the boreal forest, Isle Royale, Michigan, USA. Ecology 73:2059- 75.

Pastor J, Naiman RT, Dewey B, McInnes P. 1988. Moose, microbes and the boreal forest. BioScience 38:770-77.

Pastor J, Dewey B, Naiman RJ, McInnes PF, Cohen Y. 1993. Moose browing and soil fertility in the boreal forests of Isle Royale National Park. Ecology 74:467-80.

Regelin WL, Schwartz CC, Franzmann AW. 1985. Seasonal energy metabolism of adult moose. J Wildl Manage 49:388-93.

Stinson G, Kurz WA, Smyth CE, Neilson ET, Dymond CC, Metsaranta JM, Boisevenue C, Rampley GJ, Li Q, White TM, Blain D. 2011. An inventory-based analysis of Canada’s managed forest carbon dynamics, 1990 to 2008. Global Change Biology 17:2227-44.

Wilmers CC, Estes JA, Edwards M, Laidre KL, Konar B. 2012. Do trophic cascades affect the storage and flux of atmospheric carbon? An analysis for sea otters and kelp forests. Frontiers Ecol Env 10:409-15.

Wilson RW, Millero FJ, Taylor JR, Walsh PJ, Christensen V, Jennings S, Grosell M. 2009. Contribution of fish to the marine inorganic carbon cycle. Science 323: 359-62.

Zimov SA, Schuur EG, Chapin FS III. 2006. Permafrost and the global carbon budget. Science 312:1612-13.

3